Multiple data sources: Converging and diverging conceptualizations of LOTE teaching

The study, uncovered Japanese Language Other Than English (LOTE) teachers’ understandings of communicative language teaching (CLT). Using the idea of multiple data sources, the project relied on open-ended interviews, classroom observations, and LOTE teacher survey responses. The data provided answers to two research questions: 1) What are LOTE teachers’ beliefs and knowledge about (communicative) language teaching? and 2) How do LOTE teachers implement CLT in their classrooms. The multiple data sources provided information that both converged and diverged, providing insights not only into communicative language teaching, but also teachers’ views of language teaching in general. The various sources allowed a richer and deeper conceptualisation of LOTE teachers and captured nuances, subtlety, and complexity that these Japanese LOTE teachers dealt with in their daily professional lives. Such databases have much to offer researchers in dealing with understanding the many aspects of LOTE teacher education in particular and teacher education in general. INTRODUCTION, RESEARCH QUESTIONS, OVERVIEW In our efforts to improve language teaching, we have overlooked the language teacher (Savignon, 1991,p.272) There are many theoretical developments of communicative language teaching (CLT) along with policy and curriculum initiatives to promote communicative language learning of Language Other Than English (LOTE) (e.g., Berns, 1990; Canale & Swain, 1980; LoBianco, 1987; Board of Senior Secondary School Studies, 1995; Littiewood, 1981; Savignon, 1983; 1997; Schulz & Bartz, 1975; Vale, Scarino & McKay, 1991). Nonetheless, there is little known about what L0TE teachers actually understand by CLT and how they implement CLT in classrooms. As Kleinsasser and Savignon (1991) note, in the specific area of LOTE teacher education, there has been “little systematic inquiry conducted into language teacher perceptions and practices” (p. 291). Moreover, in the recent general teacher education research area, the question of how teachers learn to teach is more concerned with what teachers actually know and how that knowledge is acquired than what teachers need to know or how they can be trained (Carter, 1990; Richardson, 1994; Golombek, 1994). Current research on teaching practices should focus on teachers’ knowledge and beliefs with relation to their practices rather than effective teaching behaviours (Richardson, 1994). Therefore, it would seem worthwhile investigating how LOTE teachers view CLT and how they actually teach in classrooms. Most Australian LOTE teachers have either received training or inservices in communicative language teaching (CLT) during the last decade. National and state initiatives to develop students’ communicative abilities in LOTE are abundant (see, for instance, Board of Senior Secondary School Studies, 1995; Clyne, Jenkins, Chen, Tsokalidou, & Wallner, 1995; Queensland Department of Education, 1989; Scarino, Vale, McKay, & Clark, 1988; Vale, Scarino, & McKay, 1991). Although problems have been identified with the teaching of LOTE in the Australian context such as articulation, low proficiency levels, lack of quality inservices, good materials, and school support (e.g., Koide, 1976; Kawagoe, 1989; Kleinsasser, forthcoming), a major issue still remains, there is little known from the teachers’ Australian Journal of Teacher Education 18 Vol. 24, No. 1, 1999 perspectives what they think CLT is or how they implement it. In essence, inservice LOTE teachers, those teaching in the schools, have not been studied in any great depth. How is CLT understood in light of the fact that national and state directives urge communicative LOTE abilities? What is happening with CLT in LOTE classrooms? This paper aims to uncover a subgroup of LOTE teachers’ beliefs and knowledge about CLT in connection with their practices which have been overlooked by both researchers and policy-makers. The larger study (Sato, 1997) sought to answer four research questions seeking to find out information concerning LOTE teachers’ beliefs and knowledge, how LOTE teachers implement CLT, how LOTE teachers acquire or develop CLT, and the implications for LOTE teacher development. In this paper, specifically two of the research questions are highlighted: (1) What are LOTE. (in this particular study, Japanese) teachers’ beliefs and knowledge about (communicative) language teaching? and (2) How do they implement CLT in their classrooms? This paper reveals Japanese LOTE teachers’ beliefs and practices about language teaching and learning while also highlighting multiple data sources that provide information that converges and diverges, resulting in a more practical understanding of LOTE instruction. The application of multiple data sources to (LOTE) teacher education research is promising in providing clearer and more appropriate description of teachers and their understanding of LOTE teaching. Surprisingly, little has been discussed with regard to the mode of inquiry within such teacher education research focusing on teacher beliefs, perceptions, and thinking until more recently. Lee and Yarger (1996) claim that in order to make comprehensive investigations of teacher education acknowledging the complexities of context, studies should entail the use of multiple sources. Although the aspect of triangulation has been argued for in the wider literature concerning education inquiry (e.g., Mathison, 1988), and, qualitative inquiry supports the use of various data sources (e.g., LeCompte, Millroy, & Preissle, 1992; Denzin & Lincoln, 1994a), in the area of teacher education, it is rarely discussed. In this paper, the importance of multiple data sources will be outlined along with a brief theoretical perspective sketching the relevance of studying teacher beliefs, perceptions, and practices. Then the participants and the various multiple data sources will be presented. Findings will then be offered from the various data sources to help begin answering the two research questions. Finally, a discussion concerning the use of multiple data sources and the findings conclude the article. THE RELEVANCE OF MULTIPLE DATA SOURCES OR , TRIANGULATION AND THE STUDY OF TEACHER BELIEFS Triangulation to some means the use of three or more differing collection strategies to affirm and articulate the validity of evidence each produces (e.g., Williamson, Karp, Dalphin, and Gray, 1982). In fact, Williamson et al., urge the use of multiple measures thereby making it possible to concentrate on the point at which a series of independent, indirect, and perhaps weak indicators converge to minimise their separate errors and maximise their overall validity (see p. 82). More recently Denzin and Lincoln (1994b), however, suggest that the use of multiple data sources (or triangulation) is an alternative to validation and not a tool or a strategy of validation. “However, the use of multiple methods, or triangulation, reflects an attempt to secure an in-depth understanding of the phenomenon in question. Objective reality can never be captured” (p. 2). Moreover they contend that multiple data sources add “rigor, breadth, and depth” (p. 2) to studies. Mathison (1988) seems to concur with this perspective and argues that “the use of any single method, just like the view of any single individual, will necessarily be subjective and Australian Journal of Teacher Education Vol. 24, No. 1, 1999 19 therefore biased” (p. 14). Therefore she places value on triangulation where one constructs meaningful explanations of the results which may be inconsistent or contradictory rather than offering a single proposition. Collecting data that tap teachers’ perceptions of communicative language teaching and their behaviours in the classroom is not easy. Organisational theorists such as March and Simon (1958) suggest observing the behaviour of organisation members, interviewing members of the organisation, and examining documents that describe standard operating procedures as ways to determine the type of organizations and what people do in them. Another organisational theorist, Perrow (1986) concurs with these strategies but relays reservations about implementing them. For instance, he found observations took too much time and were costly and, in using interviews from industrial organisation workers, he questioned the extent to which the answers he received were accurate. More recently Kleinsasser (1993) investigated foreign language teachers’ construction of their organisation’s technical culture using data from interviews, observations, and surveys. He found participants in the study shared similar information across the three data sets while the data sets as a whole offered a more contextual understanding of foreign language teachers’ workplaces. Regardless of time or energy involved, the quality of multiple data sets does offer a clearer and more detailed description of that being studied. As Pajares (1992) reminds researchers of the dimensions in studying beliefs: “it is also clear that, if reasonable inferences about beliefs require assessments of what individuals say. intend, and do, then teachers’ verbal expressions, predispositions to action, and teaching behaviours must all be included in assessments of beliefs” (p. 327). It is important to emphasise that studies on teacher beliefs have been scarce (Clark & Peterson, 1986; Pajares, 1992) and have only gained prominence lately (Richardson, 1996). In an important review of an educational issue, Pajares synthesised research on beliefs and argued that “teachers’ beliefs can and should become an important focus of education inquiry” (p. 307). Pajares addressed numerous assumptions when studying teachers’ educational beliefs. Among them, he contended that beliefs help individuals define and understand the world and themselves, epistemological beliefs play a key role in knowledge interpretation and cognitive monitoring, and individuals’ beliefs strongly affect